MxT: Mamba x Transformer for Image Inpainting

TL;DR

MxT introduces a hybrid model combining Mamba and transformer modules for efficient and high-quality image inpainting, effectively capturing local textures and global context.

Contribution

The paper proposes a novel Hybrid Module that synergistically combines Mamba and transformer architectures for improved image inpainting.

Findings

Outperforms state-of-the-art methods on CelebA-HQ and Places2 datasets.

Efficient long-range data interaction with linear computational costs.

Enhances image reconstruction quality and contextual accuracy.

Abstract

Image inpainting, or image completion, is a crucial task in computer vision that aims to restore missing or damaged regions of images with semantically coherent content. This technique requires a precise balance of local texture replication and global contextual understanding to ensure the restored image integrates seamlessly with its surroundings. Traditional methods using Convolutional Neural Networks (CNNs) are effective at capturing local patterns but often struggle with broader contextual relationships due to the limited receptive fields. Recent advancements have incorporated transformers, leveraging their ability to understand global interactions. However, these methods face computational inefficiencies and struggle to maintain fine-grained details. To overcome these challenges, we introduce MxT composed of the proposed Hybrid Module (HM), which combines Mamba with the transformer in a synergistic manner. Mamba is adept at efficiently processing long sequences with linear computational costs, making it an ideal complement to the transformer for handling long-scale data interactions. Our HM facilitates dual-level interaction learning at both pixel and patch levels, greatly enhancing the model to reconstruct images with high quality and contextual accuracy. We evaluate MxT on the widely-used CelebA-HQ and Places2-standard datasets, where it consistently outperformed existing state-of-the-art methods. The code will be released: {\url{https://github.com/ChrisChen1023/MxT}}.